On the assessment of material removal rate and surface characteristics of Ti-6Al-4V processed by WEDM

Among α + β titanium alloys, Ti-6Al-4V (grade-5) due to superior corrosion resistance and higher strength-to-weight ratio with low density broadly preferred in the field of manufacturing, aerospace and as metallic biomaterial for orthopaedics implants [1, 2]. However, because of their low thermal conductivity and chemical reactivity with tool materials at elevated temperature, the machining of titanium alloys with the conventional machining techniques is quite a difficult task [3]. Therefore, advanced or non-traditional processes were extensively explored by the researchers to tackle the difficulty of handling titanium alloys by conventional machining processes [4–6].

Amid all of the non-traditional methods, wire electro discharge machining (WEDM) is a predominantly employed manufacturing technique for the machining of conductive workpieces with greater accuracy [7]. In this process, the sparks produced at successive interval of few micro-seconds generates heat between the workpiece and moving wire, causing the erosion of workpiece material due to the action of melting and evaporation [8–10]. This whole machining process is submerged and carried out in dielectric medium usually deionized water. The surface texture of WEDMed surface intensely affected by the thermal energy initiated, which leads to the surface roughness and changed surface morphology during the machining [11, 12]. Numerous machining parameters such as P-on, P-off, wire tension, wire speed, wire feed, current, coolant pressure, voltage, etc directly involves in the rate of material removal from the workpiece. Additionally, the generation of micro-cracks and craters during the machining action significantly influence the machined surface, thereby increasing the roughness [13, 14].

WEDM is frequently practiced by the researchers to analyze the quality and surface topography of machined surface by optimizing the material properties and input machining parameters [15–17]. Soni et al [18] studied the cutting rate (CR) and surface roughness (SR) of Ti₅₀Ni₄₉Co₅ using wire electro spark machining. In their study, the cutting rate and surface roughness was directly influenced with the change in pulse-on-time. In addition, more cracks on the machined surface witnessed at higher surface roughness and cutting rate. Kumar et al [19] performed experimentation on titanium grade-2 and noticed that pulse-on and discharge current were the most dominant parameters affecting the WEDM process. The rate of high discharge energy results in the formation of deep and wide craters, cracks and debris globules on the machined surface. Similar findings of pulse-on effect on surface finish of D2 and H13 tool steel was also reported by Lee and Tai [20] using WEDM process. Goswami and Kumar [21] employed WEDM practice for the investigation of surface integrity and material removal rate for Nimonic 80 A alloy. They observed that recast layer directly influenced with the intensity of pulse-on and longer P-on causes thicker recast layer. Pulse-on-time and pulse-off-time were the prominent factors with contribution of 46% and 33% respectively affecting the MRR. Goyal [22] studied the MRR and SR of Inconel 625 machined by normal and cryogenic-treated zinc coated wire. Furthermore, analysis of variance (ANOVA) confirmed the significance of pulse-on, cryogenic treated electrode and discharge current as prominent WEDM parameters.

Previous studies [23–26] witnessed that heat treatment of titanium alloys contributed in its improved microhardness, static and fatigue properties along with the better quality of microstructures and surface finish. For instance, Zhou et al [27] inspected the porosity of cold spray coated Ti-6Al-4V by varying the heat treatment temperature between 600–1100 °C. They observed that with the increase in heat treatment temperature, the coating on substrate material tends to be denser. Similar research was carried out by Syed et al [28] to investigate the mechanical properties and porosity of heat treated titanium grade-5 at 920 °C and 1030 °C processed by EBM. Kiel-Jamrozik et al [29] discussed about the processing of Ti-6Al-4V alloy using SLM technique. They noticed that SLM produced thermal stresses, which are necessitate to alter using heat treatment to obtain alloys with better microstructure and mechanical properties. It had been also reported by wang et al [30] that heat treatment of titanium alloys proliferate bone-implant adhesion by the growth of stem cells more rapidly. The in-vitro cytocompatibility analysis revealed the significance of heat treatment to tailor the physiochemical properties. Microstructure and corrosion behavior of heat treated Ti-6Al-4V was explored by Xu et al [31] via direct metal laser sintering (DMLS) technique. Five different range temperatures i.e. 450 °C, 550 °C, 650 °C, 950 °C and 1050 °C were selected with air cooling and water quench as cooling mediums. It was noticed that specimen quenched in water with 1050 °C heat treatment temperature comes out with superior corrosion resistance. Different heat treatment processes was studied by Altug et al [32] to inspect the microstructure, conductivity of zinc-coated brass wire and microhardness of Ti-6Al-4V processed by WEDM. Amongst all heat treatments, they found that heat treated titanium alloy at 1050 °C for 1 h followed by water quenching results in the superior surface roughness.

By scrutiny of the previously published research, most of the work targeted around the WEDM of titanium alloys and investigating the optimum machining conditions from the selected machining parameters. While, the machinability of heat treated Ti-6Al-4V in wire electrical discharge machining was rarely reported. Heat treatment of titanium is a foremost process to improve the surface properties and hardness, favourable for numerous applications. Therefore, this paper address, the investigations performed on heat treated Ti-6Al-4V to inspect the influence of selected input parameters on output variables i.e. material removal rate and surface roughness processed by WEDM via taguchi array approach and statistically analyzed by ANOVA. Additionally, the surface morphology of machined specimens has been viewed via scanning electron microscope (SEM) technique.

Ti-6Al-4V procured from Baoji Fuyuantong Industry and Trade Co. Ltd, China was used as workpiece material with chemical composition of Al 5.75%; Fe 0.28%; C 0.08%; V 3.9%; O 0.03%; Ti 89.96% and in the semi-circular shape of diameter 40 mm and thickness 6 mm for the current experimental study. The specimens were initially heat treated (figure 1) at 1100 °C for 1 h followed by water quenching to increase its hardness prior to wire EDM. The temperature of the muffle furnace kept on 1100 °C for 30 min rising from room temperature (25 °C approx) before the placing of specimens.

Zoom In Zoom Out Reset image size

The hardness of random heat treated specimens was checked using diamond indenter with an applied load of 150 kgf via ENKAY enterprises made Rockwell hardness tester in HRC scale and tabulated in table 1. Three readings were taken on each and an average value was considered, which comes out as 39 HRC whereas untreated specimen depicts 31 HRC. An overview representing the phasing of research work is shown in figure 2. The specimens were then considered for WEDM (Electronica, model ELPLUS 40A DLX) using brass wire of diameter 0.25 mm and having chemical composition as Cu 63.02%; Zn 36.85%; Pb 0.08%; Fe 0.05%. Brass wire is frequently used for industrial applications owing to their economical cost and tendency to withstand tensile stresses produced during the machining process. The diagrammatic view of wire electrical discharge machining process is shown in figure 3.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Taguchi's methodology of L₂₅ orthogonal array was employed to design the experimental trials (DoE) for the current study. The most influencing WEDM parameters such as pulse-on-time (P-on), pulse-off-time (P-off), current, voltage and wire feed rate each with five levels were selected which are tabularized in table 2. Based on the selected input parameters, table 3 represents level combinations along with the actual values of machining parameters. Whereas wire electrode (brass); dielectric medium (De-ionized water); flushing pressure (10 bar); polarity (straight) and wire tension (10 kg-f) were kept constant during experimentation. The physical properties of workpiece and electrode are listed in table 4 and the characteristics of dielectric medium i.e. de-ionized water is demonstrate in table 5. To minimize the noise factor and for better output results, each experimental run was performed twice and an average of both the repetitions was recorded for the further analysis of signal-to-noise ratios and statistically validation through ANOVA.

Material removal rate (MRR) and surface roughness (SR) were the output response parameters opted for the present study. The MRR of each experimental run was calculated by volume removed per unit machining time (VMMR) method using equation (1):-

$$\begin{eqnarray}&&{\rm{VMRR}}=\displaystyle \frac{{\rm{Volume}}\,{\rm{removed}}\,\left({{\rm{mm}}}^{3}\right)}{\mathrm{time}\,\left({\rm{minutes}}\right)}\end{eqnarray} \tag{ 1 }$$

A fixed volume of 150 mm³ was removed for all experimental trials in the form of a square block of thickness 6 mm (5 × 5 × 6 mm) and machining time for each was noted down. An average roughness value (Ra) of three measurements via Mitutoyo SJ-201 roughness tester was considered for the evaluation of surface quality of the machined surface. According to Taguchi-based methodology, the S/N ratios were calculated considering equations (2) and (3) with the approach of 'larger-is-better' for material removal rate and 'smaller-is-better' for surface roughness respectively. The feature with larger value represents superior machining performance, for instance material removal rate (MRR) whereas for better surface finish, roughness should be minimize and hence refereed under smaller-is-better criterion [33].

$$\begin{eqnarray}&&{\left(S/N\right)}_{Larger-is-better}=-10{\mathrm{log}}_{10}\left\{\displaystyle \frac{1}{n}\displaystyle \sum _{i=1}^{n}\left(\displaystyle \frac{1}{{{{\rm{y}}}_{i}}^{2}}\right)\right\}\end{eqnarray} \tag{ 2 }$$

$$\begin{eqnarray}&&{\left(S/N\right)}_{Smaller-is-better}=-10{\mathrm{log}}_{10}\left\{\displaystyle \frac{1}{n}\displaystyle \sum _{i=1}^{n}\left({{{\rm{y}}}_{i}}^{2}\right)\right\}\end{eqnarray} \tag{ 3 }$$

where n is the number of repetitions and y_i is the response value at ith experimental trial.

Besides, the dominance of input factors affecting the output responses were determined using S/N ratios and analysis of variance (ANOVA). The contribution of each selected parameter was calculated based upon their p-value and F-value. Furthermore, SEM micrographs were captured using Joel made, JSM-6610LV scanning electron microscope at magnification level of 500× to analyze the changed surface of machined specimens.

The detail description of output response observations and their measurements of both the repetitions in conjunction with the calculated values of S/N ratios for each experimental trial are summarized in table 6. A relative assessment of material removal rate and surface roughness of the specimens with respect to input parameters are unfolded in figures 4 and 5 respectively.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

WEDM parametric influence on the MRR

WEDM parametric influence on the surface roughness

In WEDM process, the rate of spark energy and machining time (P-on) directly involves in the material removal due to the thermal energy produced. Consequently, the increase of P-on at higher intensity of applied current leads more thermal energy transmitted to the workpiece causing rapid cutting rate and thus, the formation of craters and cracks on the finished surface, thereby resulting in increased roughness [36, 37]. On the same grounds, the machined sample depicts higher surface finish (1.301 μm, Trial 4) at low P-on 112 μs. Likewise, at highest P-on value (128 μs, Trial 23), a higher surface roughness of 3.038 μm was witnessed. Wire feed rate had an insignificant effect on surface roughness. It can be observed from figure 5 that the surface roughness also revealed alike fashion with P-on i.e. increased with the rise in pulse-on. Similar findings were exhibited by statistical analysis (ANOVA) table 8, carried out to spot the noteworthy parameters influencing the roughness of machined specimens. P-on was most dominant factor (contribution: 70.49%) for the surface finish of WEDMed Ti-6Al-4V surface followed by P-off (contribution: 13.33%) and current (contribution: 8.47%).

Table 8.  ANOVA for S/N ratios of Surface Roughness.

Source	DF	Seq SS	Adj MS	F-value	p-value	% contribution
P-on	4	71.884	17.9710	39.97	0.002a	70.49
P-off	4	13.601	3.4004	7.56	0.038b	13.33
Current	4	8.639	2.1597	4.80	0.079b	8.47
Voltage	4	4.252	1.0631	2.36	0.212	4.17
Wire feed rate	4	1.791	0.4477	1.00	0.502	1.76
Residual Error	4	1.798	0.4496			1.78
Total	24	101.966				100

^aMost significant; ^bSignificant factors.

Figures 6(a)–(d) illustrates the morphology of the untreated Ti-6Al-4V surface, machined surface with minimum (Trial 5) and maximum (Trial 21) material removal rate. The untreated surface (figure 6(a)) emphasizes the presence of flakes and multiple scratches. SEM analysis of WEDMed surface portrays that higher amount of spark energy with longer P-on (figures 6(c), (d), Trial 21) illustrates more distorted surface texture compared to high spark energy with low P-on (figure 6(b), Trial 5). It is due to the reason that thermal energy generated for longer period (high P-on) produces intense heat, melts or vaporize the material more rapidly resulting in the formation of surface irregularities [38].

Zoom In Zoom Out Reset image size

It should be noted that the cracks and size of the craters is affected by the amount of energy that is allowed to pass through each of the sparks. For the same reason, the surface with maximum value of MRR depicts the presence of evenly distributed craters (figure 6(c)) and micro-cracks (figure 6(d)). The SEM micrograph also discloses the presences of debris globules and droplets of molten metal on the finished or machined surface. In general, with the amplification of pulse duration and applied current, the density of cracks also increases more pronouncedly.

The WEDM practice has been investigated for heat treated Ti-6Al-4V with brass wire in the current experimentation. Longer P-on and lower wire feed rate exhibits superior MRR (59.021 mm³ min⁻¹, Trial 21) than that at higher wire feed rate and shorter P-on (2.973 mm³ min⁻¹, Trial 5) at the identical discharge current i.e. 230 A. It was observed that voltage is insignificant factor for the MRR. Notably, wire feed rate revealed as most dominant parameter for MRR with contribution of 48.19% followed by P-on 29.01% at 95% confidence level analyzed by ANOVA. While, P-on comes out most significant with 70.49% contribution in case of surface roughness and superior surface finish was witnessed in Trial 4 (1.301 μm). It was noticed that P-on has direct influence on surface roughness. Furthermore, the machined WEDMed surface at greater MRR (Trial 21) illustrate micro-cracks, micro-voids and droplets of molten material compared to the cracks, wide and deeper craters on the surface depicting lowest MRR (Trial 5) during the SEM investigation.

Finally, it can be concluded that WEDM is prominent technique to machine heat treated titanium alloys by further considering in different quenching mediums. Tungsten, molybdenum and coated wires can be explored as electrode for specified industrial or medical applications.